Landslide Deformation Dynamic Prediction Based on Self-memorization Discrete Model

Considering the non-linear specificity and monotonic growth characteristics of the time series of landslide deformation, a dynamic prediction method with self-memorization model of landslide deformation is established based on the dynamic data retrieved model and self-memorization equation. By treating the time series data of monitored landslide deformation as the particular solution of the nonlinear dynamic model of landslide deformation, the differential equation describing dynamic characteristics of the landslide deformation system is deduced by using the dynamic model retrieved model. Then, the differential equation is evolved into a differential deduction -- integral equation by introducing the memory function to establish a self-memorization model of the dynamic system for predicting nonlinear landslide deformation. The model is applied to predicting the deformation time series data monitored at the Gushuwu landslide and Maoping landslide. The cases show that the self- memorization model is valid and feasible in predicting deformation of landslides.